Speaker race identification from acoustic cues in the vocal signal

Updated Acoustic Normative Data through the Lifespan: A Scoping Review

OBJECTIVE

To assess the recent literature for voice acoustic data values reported for individuals without voice disorder through the lifespan as a means to develop an updated normative acoustic data resource for children and adults.

METHODS

A scoping review was conducted using the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) Checklist. English language, full-text publications were identified through Medline (EBSCO & OVID), PubMed, APA PsycINFO, Web of Science, Google Scholar, and ProQuest Theses and Dissertations Global.

RESULTS

A total of 903 sources were retrieved; of these 510 were duplicates. Abstracts of 393 were screened, with 68 full-text review. From the eligible studies, citation review yielded 51 additional resources. Twenty-eight sources were included for data extraction. For the normative acoustic data extracted for males and females across the lifespan, lower fundamental frequency for adult females was observed and few studies collected semitone range, sound level range, or frequency range. Data extraction also indicated a predominately gender binary reporting of acoustic measures with few studies reporting gender identity, race, or ethnicity as variables of interest.

CONCLUSIONS

The scoping review yielded updated acoustic normative data that is of value for clinicians and researchers who rely on this normative data to make determinations about vocal function. The limited availability of acoustic data by gender, race, and ethnicity creates barriers for generalization of these normative values across all patients, clients, and research volunteers.

Variation in global and intonational pitch settings among black and white speakers of Southern American English

This article revisits classic questions about how pitch varies between groups by examining global and intonational pitch differences between black and white speakers from Memphis, Tennessee, using data from read speech to control for stylistic and segmental variables. Results from both mixed-effects regression modeling and smoothing spline analysis of variance find no difference between black and white men in mean F₀ and pitch range measures. However, black women produced consistently lower mean F₀ than white women. These findings suggest that while pitch patterns in black women's speech remain underexplored in the literature, they may play an important role in shaping attitudes and ideological associations concerning black American speakers in general. Moreover, vocal pitch may be a linguistic variable subject to variation, especially in a context of racialized and gendered linguistic standards.

The Effect of Mandarin Vowels on Acoustic Analysis: A Prospective Observational Study

OBJECTIVES

Although vowels are of interest for acoustic analysis in clinics, there is no consensus regarding the effect of vowel selection on acoustic perturbation parameters. This study aimed to reveal the effects of Mandarin vowels on acoustic measurements.

STUDY DESIGN

A prospective observational study.

METHODS

This prospective observational study enrolled normal phonation Mandarin speakers at the Otolaryngology Department of the Eye & ENT Hospital affiliated with Fudan University from December 2020 to August 2021. This study recruited 107 normal-voiced Mandarin speakers (59 women and 49 men) with a median age of 26 (22, 33) years old. The objective measures included traditional acoustic parameters (fundamental frequency, harmonic-to-noise ratio, percent jitter, and percent shimmer) and cepstral analysis (smoothed cepstral peak prominence) of six Mandarin vowels (ɑ /a/, o /o/, e /ɤ/, i /i/, u /u/, ü /y/).

RESULTS

The acoustic analysis revealed no significant differences in the fundamental frequency among vowels. The low vowel /a/ had the highest values for percent jitter and percent shimmer and the lowest harmonic-to-noise ratio value. The back vowel /u/ had the lowest cepstral measures (P < 0.05).

CONCLUSIONS

The acoustic analysis significantly varied across the different Mandarin vowels, and these differences must be considered for the effective clinical application of objective evaluations.

Perception of group membership from spontaneous and volitional laughter

Laughter is a ubiquitous social signal. Recent work has highlighted distinctions between spontaneous and volitional laughter, which differ in terms of both production mechanisms and perceptual features. Here, we test listeners' ability to infer group identity from volitional and spontaneous laughter, as well as the perceived positivity of these laughs across cultures. Dutch (n = 273) and Japanese (n = 131) participants listened to decontextualized laughter clips and judged (i) whether the laughing person was from their cultural in-group or an out-group; and (ii) whether they thought the laughter was produced spontaneously or volitionally. They also rated the positivity of each laughter clip. Using frequentist and Bayesian analyses, we show that listeners were able to infer group membership from both spontaneous and volitional laughter, and that performance was equivalent for both types of laughter. Spontaneous laughter was rated as more positive than volitional laughter across the two cultures, and in-group laughs were perceived as more positive than out-group laughs by Dutch but not Japanese listeners. Our results demonstrate that both spontaneous and volitional laughter can be used by listeners to infer laughers' cultural group identity. This article is part of the theme issue 'Voice modulation: from origin and mechanism to social impact (Part II)'.

"Sounding Black": Speech Stereotypicality Activates Racial Stereotypes and Expectations About Appearance

Black Americans who are perceived as more racially phenotypical-that is, who possess more physical traits that are closely associated with their race-are more often associated with racial stereotypes. These stereotypes, including assumptions about criminality, can influence how Black Americans are treated by the legal system. However, it is unclear whether other forms of racial stereotypicality, such as a person's way of speaking, also activate stereotypes about Black Americans. We investigated the links between speech stereotypicality and racial stereotypes (Experiment 1) and racial phenotype bias (Experiment 2). In Experiment 1, participants listened to audio recordings of Black speakers and rated how stereotypical they found the speaker, the likely race and nationality of the speaker, and indicated which adjectives the average person would likely associate with this speaker. In Experiment 2, participants listened to recordings of weakly or strongly stereotypical Black American speakers and indicated which of two faces (either weakly or strongly phenotypical) was more likely to be the speaker's. We found that speakers whose voices were rated as more highly stereotypical for Black Americans were more likely to be associated with stereotypes about Black Americans (Experiment 1) and with more stereotypically Black faces (Experiment 2). These findings indicate that speech stereotypicality activates racial stereotypes as well as expectations about the stereotypicality of an individual's appearance. As a result, the activation of stereotypes based on speech may lead to bias in suspect descriptions or eyewitness identifications.

Volume and Verve: Understanding Correction/Behavioral Warnings in Teacher-Child Classroom Interactions Involving an African American Kindergarten Student

Purpose Using a framework of culturally and linguistically responsive classroom management, this study employed both quantitative and qualitative analyses of ethnographic data to illustrate patterns of corrections/behavioral warnings in teacher-child interactions for an African American child in the classroom. Method Three child participants and their teacher were recruited for the study: MJ, an African American kindergartner, the focal child, and two same-sex Filipino classmates who spoke a nonmainstream dialect of English, Ben and Leo, for frame of reference. None of the three boys were diagnosed with language-learning difficulties. Data analyses included (a) categorical analyses of observational field notes taken across a 7-week period of classroom observation and (b) situated discourse analysis taken from video-recorded small group literacy lessons also in the classroom. Results Two key findings emerged. First, MJ, the focal participant, received a relatively high frequency of correction/behavioral warnings, both relative to the other forms of teacher-initiated interaction and also relative to his two classmates. Second, the majority of MJ's corrections/behavioral warnings were directed toward his volume and verve-features that have been associated with the communication style of many African American students. Conclusions We need to include teachers and administrators in our discussions about the communication style of African American students and broaden these discussions to explicitly consider the influence of nonverbal features, such as volume and verve, on patterns of teacher-student communication interactions. In particular, such communication features may be contributing to high-stakes outcomes for African American children, such as referrals, diagnoses, educational placements, and disciplinary actions.

Face-voice space: Integrating visual and auditory cues in judgments of person distinctiveness

Faces and voices each convey multiple cues enabling us to tell people apart. Research on face and voice distinctiveness commonly utilizes multidimensional space to represent these complex, perceptual abilities. We extend this framework to examine how a combined face-voice space would relate to its constituent face and voice spaces. Participants rated videos of speakers for their dissimilarity in face only, voice only, and face-voice together conditions. Multiple dimensional scaling (MDS) and regression analyses showed that whereas face-voice space more closely resembled face space, indicating visual dominance, face-voice distinctiveness was best characterized by a multiplicative integration of face-only and voice-only distinctiveness, indicating that auditory and visual cues are used interactively in person-distinctiveness judgments. Further, the multiplicative integration could not be explained by the small correlation found between face-only and voice-only distinctiveness. As an exploratory analysis, we next identified auditory and visual features that correlated with the dimensions in the MDS solutions. Features pertaining to facial width, lip movement, spectral centroid, fundamental frequency, and loudness variation were identified as important features in face-voice space. We discuss the implications of our findings in terms of person perception, recognition, and face-voice matching abilities.

Sociophonetic trends in studies of Southern U.S. English

Although the U.S. South was usually avoided for phonetic studies in earlier years, phonetic analyses of Southern U.S. English have expanded in numerous directions in recent years. Studies of vocalic variation have dominated the enterprise, with a broad range of studies that have examined the phonetic peculiarities and distribution of variants within the region, the characteristics of particular communities, and the phonetic attributes of African American and Latino groups. However, other phonetic phenomena are finally seeing more study. Several consonantal variables have attracted acoustic analysis. Other studies have focused on intonation and certain timing-related phenomena. Numerous experimental studies have tested the abilities of listeners to identify dialects and a few other perceptual issues. These developments are outlined here, including advancements in segmental, prosodic, and perceptual analyses.

Acoustic Voice Modifications in Individuals with Parkinson Disease Submitted to Deep Brain Stimulation

Introduction  Subthalamic nucleus deep brain stimulation (STN-DBS) improves motor function in individuals with Parkinson disease (PD). The evidence about the effects of STN-DBS on the voice is still inconclusive. Objective  To verify the effect of STN-DBS on the voice of Brazilian individuals with PD. Methods  Sixteen participants were evaluated on the Unified Parkinson Disease Rating Scale-Part III, and by the measurement of the acoustic modifications in on and off conditions of stimulation. Results  The motor symptoms showed significant improvement with STN-DBS on. Regarding the acoustic measures of the voice, only the maximum fundamental frequency (fhi) showed a statistical difference between on- and off-conditions, with reduction in off-condition. Conclusion  Changes in computerized acoustic measures are more valuable when interpreted in conjunction with changes in other measures. The single finding in fhi suggests that DBS-STN increases vocal instability. The interpretation of this result should be done carefully, since it may not be of great value if other measures that also indicate instability are not significantly different.

Telling Friend from Foe: Listeners Are Unable to Identify In-Group and Out-Group Members from Heard Laughter

Group membership is important for how we perceive others, but although perceivers can accurately infer group membership from facial expressions and spoken language, it is not clear whether listeners can identify in- and out-group members from non-verbal vocalizations. In the current study, we examined perceivers' ability to identify group membership from non-verbal vocalizations of laughter, testing the following predictions: (1) listeners can distinguish between laughter from different nationalities and (2) between laughter from their in-group, a close out-group, and a distant out-group, and (3) greater exposure to laughter from members of other cultural groups is associated with better performance. Listeners (n = 814) took part in an online forced-choice classification task in which they were asked to judge the origin of 24 laughter segments. The responses were analyzed using frequentist and Bayesian statistical analyses. Both kinds of analyses showed that listeners were unable to accurately identify group identity from laughter. Furthermore, exposure did not affect performance. These results provide a strong and clear demonstration that group identity cannot be inferred from laughter.

Development of the Arabic Voice Pathology Database and Its Evaluation by Using Speech Features and Machine Learning Algorithms

A voice disorder database is an essential element in doing research on automatic voice disorder detection and classification. Ethnicity affects the voice characteristics of a person, and so it is necessary to develop a database by collecting the voice samples of the targeted ethnic group. This will enhance the chances of arriving at a global solution for the accurate and reliable diagnosis of voice disorders by understanding the characteristics of a local group. Motivated by such idea, an Arabic voice pathology database (AVPD) is designed and developed in this study by recording three vowels, running speech, and isolated words. For each recorded samples, the perceptual severity is also provided which is a unique aspect of the AVPD. During the development of the AVPD, the shortcomings of different voice disorder databases were identified so that they could be avoided in the AVPD. In addition, the AVPD is evaluated by using six different types of speech features and four types of machine learning algorithms. The results of detection and classification of voice disorders obtained with the sustained vowel and the running speech are also compared with the results of an English-language disorder database, the Massachusetts Eye and Ear Infirmary (MEEI) database.

Your Cheatin' Voice Will Tell on You: Detection of Past Infidelity from Voice

Evidence suggests that many physical, behavioral, and trait qualities can be detected solely from the sound of a person's voice, irrespective of the semantic information conveyed through speech. This study examined whether raters could accurately assess the likelihood that a person has cheated on committed, romantic partners simply by hearing the speaker's voice. Independent raters heard voice samples of individuals who self-reported that they either cheated or had never cheated on their romantic partners. To control for aspects that may clue a listener to the speaker's mate value, we used voice samples that did not differ between these groups for voice attractiveness, age, voice pitch, and other acoustic measures. We found that participants indeed rated the voices of those who had a history of cheating as more likely to cheat. Male speakers were given higher ratings for cheating, while female raters were more likely to ascribe the likelihood to cheat to speakers. Additionally, we manipulated the pitch of the voice samples, and for both sexes, the lower pitched versions were consistently rated to be from those who were more likely to have cheated. Regardless of the pitch manipulation, speakers were able to assess actual history of infidelity; the one exception was that men's accuracy decreased when judging women whose voices were lowered. These findings expand upon the idea that the human voice may be of value as a cheater detection tool and very thin slices of vocal information are all that is needed to make certain assessments about others.

One's Voice: A Central Component of Personal Factors in Augmentative and Alternative Communication

Augmentative and alternative communication (AAC) devices have opened the gates to interaction for those with severe communication impairments. In the assessment and intervention, all components of the World Health Organization's International Classification of Functioning, Disability, and Health (ICF) should be addressed. However, an important Personal Factor to full integration has been largely ignored—that of one's voice. Each one of us has a unique voice that conveys our age, cultural background and personality—it's how people know and remember you. These affordances of the natural voice are not available to those who express themselves using AAC devices. A personalized digital voice brings the field of AAC to a closer realization of the social model of disability in which individuals are not defined by their disability and it is just one aspect of them. Access to a personalized voice uplifts the AAC user and provides an opportunity for social and emotional engagement that enhances quality of life.

A comparison of vocal parameters in adult bilingual Hebrew-English speakers

PURPOSE

There has been growing research on the effects of language on voice characteristics; however, few studies have examined the impact of language on vocal features within bilinguals. This study aimed to compare vocal parameters among bilingual Hebrew/English speaking individuals when speaking in Hebrew versus English.

METHODS

Forty bilingual participants (17 males and 23 females) between the ages of 23-60 years were asked to spontaneously speak about a neutral topic. Voice samples were digitalized into a tablet for perceptual and acoustic analyses of selected parameters.

RESULTS

Results show that there are changes in resonance, glottal attack, fundamental frequency variation and speech rate when adult bilingual speakers talk in Hebrew as compared to English.

CONCLUSION

These findings provide evidence that language plays a role in affecting vocal characteristics of bilingual individuals when they speak different languages.

Judgments of self-identified gay and heterosexual male speakers: Which phonemes are most salient in determining sexual orientation?

While numerous studies have demonstrated that a male speaker's sexual orientation can be identified from relatively long passages of speech, few studies have evaluated whether listeners can determine sexual orientation when presented with word-length stimuli. If listeners are able to distinguish between self-identified gay and heterosexual male speakers of American English, it is unclear whether they form their judgments based on a phoneme, such as a vowel or consonant, or multiple phonemes, such as a vowel and a consonant. In this study, we first found that listeners can distinguish between self-identified gay and heterosexual speakers of American English upon hearing word-length stimuli. We extended these results in a separate experiment to demonstrate that listeners primarily rely on vowels, and to some extent consonants, when forming their judgments. Listeners were able to differentiate between the two groups of speakers for each of the vowels and three of the seven consonants presented. In a follow-up experiment we found evidence that listeners' judgments improved if they were presented with multiple phonemes, such as a vowel and /s/. These results provide important information about how different phonemes can provide discriminant information about a male speaker's sexual orientation.

Sounding Black or White: priming identity and biracial speech

Research has shown that priming one’s racial identity can alter a biracial individuals’ social behavior, but can such priming also influence their speech? Language is often used as a marker of one’s social group membership and studies have shown that social context can affect the style of language that a person chooses to use, but this work has yet to be extended to the biracial population. Audio clips were extracted from a previous study involving biracial Black/White participants who had either their Black or White racial identity primed. Condition-blind coders rated Black-primed biracial participants as sounding significantly more Black and White-primed biracial participants as sounding significantly more White, both when listening to whole (Study 1a) and thin-sliced (Study 1b) clips. Further linguistic analyses (Studies 2a–c) were inconclusive regarding the features that differed between the two groups. Future directions regarding the need to investigate the intersections between social identity priming and language behavior with a biracial lens are discussed.

Talker identification across source mechanisms: experiments with laryngeal and electrolarynx speech

PURPOSE

The purpose of this study was to determine listeners' ability to learn talker identity from speech produced with an electrolarynx, explore source and filter differentiation in talker identification, and describe acoustic-phonetic changes associated with electrolarynx use.

METHOD

Healthy adult control listeners learned to identify talkers from speech recordings produced using talkers' normal laryngeal vocal source or an electrolarynx. Listeners' abilities to identify talkers from the trained vocal source (Experiment 1) and generalize this knowledge to the untrained source (Experiment 2) were assessed. Acoustic-phonetic measurements of spectral differences between source mechanisms were performed. Additional listeners attempted to match recordings from different source mechanisms to a single talker (Experiment 3).

RESULTS

Listeners successfully learned talker identity from electrolarynx speech but less accurately than from laryngeal speech. Listeners were unable to generalize talker identity to the untrained source mechanism. Electrolarynx use resulted in vowels with higher F1 frequencies compared with laryngeal speech. Listeners matched recordings from different sources to a single talker better than chance.

CONCLUSIONS

Electrolarynx speech, although lacking individual differences in voice quality, nevertheless conveys sufficient indexical information related to the vocal filter and articulation for listeners to identify individual talkers. Psychologically, perception of talker identity arises from a "gestalt" of the vocal source and filter.

Towards a more nuanced view of vocal attractiveness

This study reports on male and female Californians' ratings of vocal attractiveness for 30 male and 30 female voices reading isolated words. While ratings by both sexes were highly correlated, males generally rated fellow males as less attractive than females did, but both females and males had similar ratings of female voices. Detailed acoustic analyses of multiple parameters followed by principal component analyses on vowel and voice quality measures were conducted. Relevant principal components, along with additional independent acoustic measures, were entered into regression models to assess which acoustic properties predict attractiveness ratings. These models suggest that a constellation of acoustic features which indicate apparent talker size and conformity to community speech norms contribute to perceived vocal attractiveness. These results suggest that judgments of vocal attractiveness are more complex than previously described.

The role of motivation and cultural dialects in the in-group advantage for emotional vocalizations

It is well-established that non-verbal emotional communication via both facial and vocal information is more accurate when expresser and perceiver are from the same cultural group. Two accounts have been put forward to explain this finding: According to the dialect theory, culture-specific learning modulates the largely cross-culturally consistent expressions of emotions. Consequently, within-group signaling benefits from a better match of the “emotion dialect” of the expresser and perceiver. However, it has been proposed that the in-group advantage in emotion recognition could instead arise from motivational differences in the perceiver, with perceivers being more motivated when decoding signals from members of their own group. Two experiments addressed predictions from these accounts. Experiment 1 tested whether perceivers' ability to accurately judge the origin of emotional signals predicts the in-group advantage. For perceived group membership to affect the perceivers' motivation, they must be able to detect whether the signal is coming from an in-group or out-group member. Although an in-group advantage was found for in-group compared to out-group vocalizations, listeners were unable to reliably infer the group membership of the vocalizer. This result indicates that improved recognition of in-group signals can occur also when the perceiver is unable to judge whether signals were produced by in- or out-group members. Experiment 2 examined the effects of expected and actual group membership of signals on emotion recognition by manipulating both orthogonally. The actual origin of the stimulus was found to significantly affect emotion recognition, but the believed origin of the stimulus did not. Together these results support the notion that the in-group advantage is caused by culture-specific modulations of non-verbal expressions of emotions, rather than motivational factors.

Vocal indices of stress: a review

OBJECTIVE

Identification of stress patterns in the voice has multiple potential applications. The objective was to review literature pertaining to the effects of various forms of stress upon the healthy voice.

STUDY DESIGN

Literature review, discussion of results, and direction for further study.

METHODS

This review article offers a model of stress and a review of the historical and recent research into the effects of stress on the voice. Electronic databases were searched using the key words. No studies were excluded on the basis of design; however, an attempt was made to include in the discussion studies which primarily address physiological and acoustic vocal parameters. The results of greater than 50 studies examining the effect of stressors ranging from lie and guilt to high altitude and space flight upon the voice were included in the review.

RESULTS

Increase in fundamental frequency is the most commonly reported effect of stress in well-controlled trials. The trend, however, is not universal. A reduction in noise as reflected by the diminished vocal jitter is reported, but less frequently.

CONCLUSIONS

Stress types, gender, and individual differences in baseline autonomic tone may explain the primarily equivocal findings of effects of stressor exposure or perceived stress on voice; and as such, the article concludes with a discussion of directions for future study.

Voice low tone to high tone ratio, nasalance, and nasality ratings in connected speech of native Mandarin speakers: a pilot study

OBJECTIVE

Voice low tone to high tone ratio (VLHR) is defined as the power ratio of a voice spectrum with a specific cut-off frequency. Previous studies have shown that there are significant correlations between VLHR and nasalance and hypernasality ratings in vowels. The correlation was investigated in this study using connected speech material.

DESIGN

The Zoo Passage, the Rainbow Passage, the English Nasal Sentences, the Mandarin Nonnasal Sentences, and the Mandarin Nasal Sentences were used to acquire VLHRs, nasalance scores, and perceptual judgments of nasality. Each passage was recorded twice for averaging, and the cut-off frequencies from 200 Hz to 1200 Hz were used to survey for the presence of optimal correlations with VLHR. Participants : Ten native Mandarin speakers with an English learning history of over 8 years were enrolled. Main outcome measures : VLHRs, nasalance scores, and hypernasality ratings.

RESULTS

The correlations of VLHR with nasalance (rho  =  .76, p < .001, Spearman rank correlation) and nasality ratings (rho  =  .81, p < .001) were significant using a cut-off frequency of 300 Hz for the English passages. For the Mandarin Sentences, the optimal correlations of VLHR with nasalance (rho  =  .83, p < .001) and nasality ratings (rho  =  .79, p < .001) were identified using a cut-off frequency of 500-Hz.

CONCLUSION

The significant correlations of VLHR with nasalance and perceptual ratings of nasality using connected speech show that these approaches have a potential value in terms of basic and clinical application.

Objective voice analysis of boys with profound hearing loss

OBJECTIVES

A critical factor that affects human voice production is hearing because it provides necessary feedback for control of speech. Vocal quality of profoundly hearing-impaired children is often considered deviant from both perceptual and acoustic perspectives. The present study compares selected vocal acoustic parameters of a profound hearing loss group of boys with normal peers.

METHODS

The subjects were 15 Iranian boys with profound bilateral sensorineural hearing loss and 15 Iranian normal hearing participants matched according to age and sex. The age range of the children with hearing loss was 61-81 months (M=72.26) (ie, 5.1-6.75 years, M=6.02 years) and for the normal group the age range was 61-80 months (M=71.47) (5.08-6.67 years, M=5.96 years). Each subject phonated 10 /â/ vowels with constant pitch and loudness for maximal phonation times. The mid 3-second portion of each token was analyzed using Dr. Speech 4.3u software (subprogram: Vocal Assessment; Dr. Speech, Tiger Electronics, Seattle, WA).

RESULTS

There was a statistically significantly higher fundamental frequency (F0), jitter, and shimmer in the productions of the hearing loss boys compared with the normal hearing boys. Consistent with these findings was a significantly lower value for the harmonics-to-noise ratio measure for the boys with hearing loss.

CONCLUSIONS

The results of the present study suggest that profoundly deaf children present with greater phonatory instability and spectral noise, with the possible inference of reduced laryngeal control relative to vocal quality. The finding of higher F0 for the boys with profound hearing loss suggests that they use a different control strategy for pitch, an area needing further study. These findings of acoustic and F0 differences of the hearing-impaired boys should be kept in mind for intervention practices especially when the social impact of deafness is considered.

Voice pitch characteristics of Cantonese and English produced by Cantonese-English bilingual children

Two-minute spontaneous conversational speech of English and Cantonese was obtained from 86 Cantonese-English bilingual children. As a physical measurement of the rate of vocal fold vibration during speech production, fundamental frequency (F0) characteristics including the mean speaking F0, pitch sigma, minimum and maximum F0 values were measured from the medial 60-second portion of the speech samples by using Praat. F0 data measured from the English and Cantonese productions were statistically compared. Significant correlation was found for speaking F0, pitch sigma, and F0 range values between Cantonese and English. Results indicated that speaking F0 and F0 range values were significantly lower in Cantonese than in English. It is speculated that such difference is related to the tonal nature of Cantonese, although language proficiency may play a role in the discrepant F0 findings.

Developmental craniofacial anthropometry: Assessment of race effects

Differences in craniofacial anatomy among racial groups have been documented in a variety of structures, but the oral and maxillofacial regions have been shown to be a particularly defining region of variability between different racial/ethnic groups. Such comparisons are informative, but they neither address developmental changes of the craniofacial anatomy nor do they assess or take into account the natural variability within individual races that may account for similar reported, across-group variations. The purpose of this report was to compare-using medical imaging studies-the growth trend of select race-sensitive craniofacial variables in the oral and pharyngeal regions when all races [White, Asian, Black, and Hispanic (AR)] are included versus only a single race category [White (WR)]. Race effect was tested by comparing sex-specific growth fits (fourth degree polynomial model) for AR versus WR data. Findings indicate that the inclusion of all races versus a single race did not significantly alter the growth model fits. Thus, the inclusion of all races permits the advancement of general growth models; however, methodologically, it is best to treat the race variable as a covariate in all future analysis to test for both potential all race effects or individual race effects, on general growth models.

Fundamental Frequency in Monolingual English, Bilingual English/Russian, and Bilingual English/Cantonese Young Adult Women

Mean F0 of nine young adult English/Russian female bilinguals and nine young adult English/Cantonese female bilinguals were examined from samples of connected speech in each language. Mean F0 were compared in each language and in English with those of a monolingual English control group of ten young adult female speakers. Acoustic measurements were analyzed with the Kay Elemetrics Multispeech program (Kay Elemetrics, Lincoln Park, NJ). The results indicate that the English/Russian bilinguals consistently had a higher mean F0 in Russian than in English. Mean F0 did not change with language switch for the English/Cantonese speakers. There were no significant differences between the groups in their English production. Clinical implications regarding norms for both monolingual and bilingual persons, as well as implications for understanding the nature of bilingualism, in particular code-switching, are discussed.

Something in the way she moves

A recent study using a crossmodal matching task showed that the identity of a talker could be recognized even when the auditory and visual stimuli that were being matched were different sentences spoken by the talker. This finding implies that general temporal features of a person's speech are shared across the auditory and visual modalities.

Estimation of speakers' weight and height from speech: a re-analysis of data from multiple studies by Lass and colleagues

Research on identification of speakers' weight and height from their speech has yielded controversial results. Conclusions from a series of reports conducted by Lass and colleagues must be interpreted with caution because they are based on comparisons of only two types of data, mean of actual values and mean of estimated values. Cohen and others clearly demonstrated that this method of analysis overstated the accuracy of listeners' judgments. Following Van Dommelen's procedure, data from six of Lass's reports were re-analyzed with more appropriate statistics. Results of re-analysis show that (a) listeners are not very efficient guessing the weight or height of speakers, as only 14%, of the estimations correlated significantly with actual values. (b) Judgments are highly consistent under different acoustic conditions, suggesting that listeners follow vocal stereotypes about the body size of speakers, even though these stereotypes are wrong.

Harmonics-to-noise ratio: an index of vocal aging

Distinguishing between vocal changes that occur with normal aging and those that are associated with disease is an important goal of research in voice. Several acoustic measures have been used in an attempt to illuminate the integrity of the vocal mechanism, including harmonics-to-noise ratio (HNR), jitter, and fundamental frequency (F0). HNR is a measure that quantifies the amount of additive noise in the voice signal; jitter reflects the periodicity of vocal fold vibration. In this study, measures of HNR, jitter and F0 were used to compare vocal function in three groups of normally speaking women: young adults, middle-aged adults, and elderly adults. Significant differences in HNR emerged between the elderly women and the other two groups. F0 differences were also apparent between the elderly group and the two younger groups; there were no significant differences in jitter between the three groups. HNRwas found to be a more sensitive index of vocal function than jitter. The significant lowering of HNR evident in t he elderly speakers may be attributable in part to medications taken by the majority of these elderly subjects.

[Acoustic voice analysis: reliability of a set of multi-dimensional parameters]

A correlational study of the reliability of acoustic voice parameters was made of 148 healthy adults. Acoustic analysis was performed with MDVP-Multi-Dimensional Voice Program implemented in a CSL-Computerized Speech Lab of Kay Elemetrics. A set of 29 voice parameters were obtained from two samples of sustained vowel /a/ recorded from each subject. General results separated by sex are showed and the test-retest reliability in each pair of measures was calculated. Data show a high intra-subject stability of Frequency Fundamental parameters; acceptable stability in parameters of Frequency and Amplitude Perturbation, Noise, Subharmonics and Voice irregularities; and a very low consistency in Tremor parameters. Parameters related with shimmer were more reliable than parameters related with jitter. According to results several conclusions are reported.

Multimodal standardization of voice among four multicultural populations: fundamental frequency and spectral characteristics

A stratified random sample of 20 males and 20 females matched for physiological factors and cultural-linguistic markers were examined to determine differences in fundamental frequency and spectral characteristics during prolongation of three vowels: [a], [i], and [u]. The ethnic-gender breakdown included four sets of five male and five female subjects comprised of Caucasian and African-American speakers of standard American English, native Hindi Indian speakers, and native Mandarin Chinese speakers. Acoustic measures were analyzed using the Multidimensional Voice Program (Kay Elemetrics, Lincoln Park, NJ) (Model 4305) from which fundamental frequency and associated acoustic spectra were extracted from a 200-ms sample of each vowel token. Statistically significant group differences for the main effects of culture, race, and gender were found. The acoustic differences found are attributed to biomechanical, physiological, cultural, and linguistic factors.

Multimodal Standardization of Voice Among Four Multicultural Populations Formant Structures

A stratified random sample of 20 males and 20 females matched for physiologic factors and cultural-linguistic markers was examined to determine differences in formant frequencies during prolongation of three vowels: [a], [i], and [u]. The ethnic and gender breakdown included four sets of 5 male and 5 female subjects comprised of Caucasian and African American speakers of Standard American English, native Hindi Indian speakers, and native Mandarin Chinese speakers. Acoustic measures were analyzed using the Computerized Speech Lab (4300B) from which formant histories were extracted from a 200-ms sample of each vowel token to obtain first formant (F1), second formant (F2), and third formant (F3) frequencies. Significant group differences for the main effect of culture and race were found. For the main effect gender, sexual dimorphism in vowel formants was evidenced for all cultures and races across all three vowels. The acoustic differences found are attributed to cultural-linguistic factors.

Men's voices and women's choices

I investigated the relationship between male human vocal characteristics and female judgements about the speaker. Thirty-four males were recorded uttering five vowels and measures were taken, from power spectrums, of the first five harmonic frequencies, overall peak frequency and formant frequencies (emphasized, resonance, frequencies within the vowel). Male body measures were also taken (age, weight, height, and hip and shoulder width) and the men were asked whether they had chest hair. The recordings were then played to female judges, who were asked to rate the males' attractiveness, age, weight and height, and to estimate the muscularity of the speaker and whether he had a hairy chest. Men with voices in which there were closely spaced, low-frequency harmonics were judged as being more attractive, older and heavier, more likely to have a hairy chest and of a more muscular body type. There was no relationship between any vocal and body characteristic. The judges' estimates were incorrect except for weight. They showed extremely strong agreement on all judgements. The results imply that there could be sexual selection through female choice for male vocal characteristics, deeper voices being preferred. However, the function of the preference is unclear given that the estimates were generally incorrect. Copyright 2000 The Association for the Study of Animal Behaviour.

Acoustic studies of dysarthric speech: methods, progress, and potential

(1) The reader will be able to describe the major types of acoustic analysis available for the study of speech, (2) specify the components needed for a modern speech analysis laboratory, including equipment for recording and analysis, and (3) list possible measurements for various aspects of phonation, articulation and resonance, as they might be manifest in neurologically disordered speech.

Aerodynamic and acoustic characteristics of the adult African American voice

Laryngeal aerodynamic and acoustic characteristics of African American voice production were examined from vowel samples produced by ten adult female and ten adult male speakers. The data were compared with that for a control group consisting of ten adult female and ten adult male White speakers, matched for age, height, and weight. All measures were analyzed using Cspeech 4.0. Aerodynamic measurements, extracted from a glottal airflow waveform, included maximum flow declination rate, alternating glottal airflow, minimum glottal airflow, and airflow open quotient. Acoustic measures included fundamental frequency and sound pressure level. No significant mean differences between the African American and White speakers were found, except for maximum-flow declination rate. The White speakers produced significantly higher declination rates than the African American speakers. The factor of sex for the African American speakers was statistically significant for the measures of maximum-flow declination rate, alternating glottal airflow, open quotient, and fundamental frequency, consistent with the functioning of the White speakers. The results suggest that during vowel production, where the vocal tract is in a fairly static position, acoustic and aerodynamic characteristics for African American and White Speakers are comparable.

A preliminary investigation of the effects of gender and race on Voice Onset Time

Twenty individuals participated in a study of Voice Onset Time (VOT) production. Participants included equal numbers of males and females and equal numbers of African Americans and Caucasian Americans. Each individual read a set of stimuli formed from the six stop consonants (/p/,/t/,/k/;/b/,/d/,/g/) combined with the three vowels /i/,/a/, and /u/. Their productions were measured for VOT. Considerably more prevoicing (i.e., negative VOT) for voiced stops was found in the present study in comparison with past studies. Statistically significant differences were found for both gender and race. These results suggest that the normative data presently available is probably inadequate because it does not accurately reflect the normal distribution of either gender or race within the American population.

Speaking fundamental frequency characteristics of 8- through 10-year-old white- and African-American boys

The purpose of this study was to determine the typical speaking fundamental frequencies (SFF) and the standard deviations of the SFF (pitch sigmas) of African-American boys in comparison to white-American boys. A group of 90 boys, 45 from each racial group, aged 8 through 10 years, were recorded as they read a passage and as they described a picture. Analysis of these recordings indicated that no significant differences occurred among any of the groups for modal SFF. The across-race analysis of pitch sigma revealed that the 9- and 10-year-old African-American boys exhibited significantly greater variability in comparison to the age-matched white boys. When the pitch sigma data were analyzed within race, the 10-year-old African-American boys exhibited significantly greater variability than the younger African-American boys, while no differences were found among the white boys.